David SpiegelhalterIlan Goodman

Professor Sir David Spiegelhalter runs about half an hour late to our meeting, and eventually appears in his slippers, apologising profusely and explaining that he had been phoned up last-minute to speak on Radio 4 about the controversy surrounding the AstraZeneca vaccine. I’ve heard worse excuses. He dashes off, gets changed, and returns, offering a cup of tea and further apologies. We take a seat on a bench outside his house – two metres apart, of course.

Professor Spiegelhalter has been a man in high demand over the past year. He has featured repeatedly on TV news, radio shows and podcasts (as well as hosting his own); written a series of articles for The Guardian; and helped to design slides for the government Covid daily briefings. If there can be such a thing as a celebrity statistician, it is him.

“It is extraordinary,” he says of his high profile. “This is very odd for a statistician, because statisticians tend to be quite retiring people behind the scenes, partly because they don’t tend to have strong agendas about what should be done. Statisticians, as a profession, just want to use data to learn more about the world and improve the kind of decisions that are made, to improve the way stuff is talked about in society – that’s what we do. So, we tend to stay in the background.”

In this regard, Spiegelhalter is a stark outlier. His career, even before the pandemic, has been characterized by an enthusiasm to talk and write publicly about his work, and a dedication to improving the way in which science and numbers are talked about and communicated. As well as having served as President of the Royal Statistical Society, he has written The Art of Statistics: How to Learn from Data; hosted the BBC’s Tails you win: the science of chance, as well as Climate Change by Numbers; and once finished seventh in an episode of Winter Wipeout, which he describes as “possibly my greatest achievement”. And he wants other statisticians to follow suit.

“I’ve kind of got a campaign to give statisticians a higher public profile, because I think that statisticians do have something quite strong to say in situations like the whole of the last year, when data has become so important. There are so many numbers in the news every day - well, what do those numbers mean? As a profession, statisticians can provide a whole lot of valuable insight.”

In Spigelhalter’s experience, this insight has been well-received. “When someone asks me ‘What do you think of this policy?’, or ‘What do you think will happen?’, I say ‘I don’t know! It’s not my job, I’m not going to say what should be done, or what’s going to happen. I don’t do that – I’m just going to interpret the data.’ The more I refuse to talk about things, the more people seem to like it, and seem to trust what I say.”

“‘Scientists say’, ‘scientists tell us’, ‘we follow the science’ – it’s complete nonsense! [...] Political decisions about what should be done are much broader than that, so the idea of ‘following the science’ is completely inappropriate”

This principle of not asserting any certainty beyond one’s own expertise is a cornerstone of trustworthy communication and one that is adopted by the Winton Centre for Risk and Evidence Communication, where Spiegelhalter is Chairman.

“We bang on relentlessly in every talk we give about trustworthy communication, which we both try to promote, and try to do ourselves. We’re trying to promote this idea as not just a nice, ethical thing to do, but something that can be actively researched.

“Our claim is that if you admit uncertainty, especially if you can be confident about your uncertainty, that doesn’t lead people not to trust you. Although we think that admitting uncertainty is just the right thing to do, it’s good to be able to tell people that if you do, it doesn’t make people trust you less. People really like it if they feel they understand something a bit better after listening to you – people love it. And I think they actually quite like it when you don’t tell them what to do, or what should happen, and don’t start wagging your finger at them, or at the government.”

On the way that science and evidence have been talked about in relation to policy, Spiegelhalter is scathing. “‘Scientists say’, ‘scientists tell us’, ‘we follow the science’ – it’s complete nonsense! I think that phrase, now, is gone. It’s awful. The last thing that politicians should be doing is just following the science. If you look at SAGE – a fine body of people, brilliantly led – they’re the first ones to admit there’s no economists on it, they have some social science input, but they’re very, very narrow in their perspective. Political decisions about what should be done are much broader than that, so the idea of ‘following the science’ is completely inappropriate.

“I don’t even like the phrase ‘evidence-based policy’. It should be evidence-informed policy, because in the end these are political decisions, and the politicians have to take responsibility for them. I always think of science not out in front, telling you where to go, but sort of mooching along beside you, muttering to itself and occasionally making comments on what it sees around it. It’s good to have there, good to have that sort of person in your team, but they shouldn’t be telling you what to do.”


Mountain View

Toby Young on cancel culture and importance of lockdown-sceptical discourse

The team at the Winton Centre had designed the slides for the government briefing on the risks of the AstraZeneca vaccine the day before our meeting, and Spiegelhalter thought the briefing had made a welcome change. “There weren’t any politicians – it was a completely science-driven group. And it was quite obvious, by the way they talked, sitting there behind a slightly shabby desk. It wasn’t all full of Union Jacks hanging in the back and all this nonsense. The contrast between that and these increasingly presidential-looking briefings in this multi-million-pound suite they’ve built in Number Ten I find really notable.”

I ask Spiegelhalter whether he feels that the reporting of science has improved over the course of the last year, and he responds by practicing what he preaches – he tells me he doesn’t know. “The trouble is I haven’t got a measure of this. I hope that this whole crisis has brought a more mature response to science, both in the media and in the public, rather than thinking of it as this monolithic body of facts. Science is contested, it’s uncertain, it doesn’t tell you what to do, and there will always be differing opinions among scientists.

“I just dread the virus going away and we’re back to ‘killer bacon sandwich’ stories. I kind of hope that having gone through something that’s so massive, important, and life-changing, that the public and media won’t go straight back to this awful ‘science’ that we get bombarded with all the time.”